The development of new micro and nanotechnologies has boosted our ability to study living systems with an unprecedent precision, with a massive increase in the volume of available experimental data imposing new challenges for theorists. It is of crucial importance to devise inference methods to extract valuable information from such data, which is typically very noisy. In this talk, I will try to show that some fundamental tools from the theory of stochastic processes and from stochastic thermodynamics may be helpful to achieve this goal. I will start by discussing how to improve the Jarzynski and Crooks free-energy estimators in single-molecule pulling experiments. Such estimators are very sensitive to experimental measurement errors and typically end up being biased. Nevertheless, they can be corrected in some simple-yet-important cases. In a second part of the talk, I intend to show that some ideas from stochastic thermodynamics can be used to assess the natural phenotypical bias between the statistics of single lineages and full cell populations. There are duality relations (à la fluctuation theorems) between single-cell and population-level distributions, which can be exploited, for instance, to infer properties of the interdivision-times distribution at the population level by using data from single lineages and viceversa. References: PRE93,032103(2016); PRE99, 042413 (2019).
Detalles de contacto:
Llorenç Serra 971 17 28 05 Contact form